Field
The present invention relates to video display, more specifically to realistic point of view video display.
Background
Two dimensional and three dimensional video systems can provide a very high quality picture. However, the image is generally “stuck” on the presentation device (e.g., screen) such that as a viewer moves across their living room, for example, the same picture is presented.
Techniques have been developed to convert existing two dimensional images into three dimensional images by manipulating a single view of a scene to generate the three dimensional image. However, there are limits as to the realism that may be present in a video generated from a single view. For example, when converting a single view, the two dimensional image may be distorted in a limited fashion. At a certain point, items occluded from the original view (e.g., no pixel information is available) are required to generate a realistic image for the scene. In some systems, the amount a viewer can move and still be presented with a realistic rendering may be limited by this occlusion distance.
Therefore, there is a need for methods and apparatus to obtain and provide realistic point of view video.